LungCT_Diagnosis

Repository processes CT scanned images of human Lungs , which are in DICOM image format. Visualises the data in 3D and trains a 3D convolution network on the data after preprocessing.

Data Acquisition

• Data can be obtained from: https://portal.gdc.cancer.gov/
• You must open with the application below-

NBIA Data Retriever (https://wiki.cancerimagingarchive.net/display/NBIA/Downloading+TCIA+Images)

Google Colab Link

The link for the google colaboratory of dicom visualisation is written is given below: (https://colab.research.google.com/drive/1y8qnK2JD8CHo90o8hpXzM1QCcoTA8vlG?usp=sharing]

The link for the google colaboratory where the entire code is written is given below: (https://colab.research.google.com/drive/1XAHBUBxucVsyv5DiiA5d3-gll-WrO3Za?usp=sharing]

The CT scanned image of a lung-

Frequency vs HU units and the corresponding .dcm image

3D images of dicom files: Segmented and rendered